Business Intelligence and Analytics: AIDS409T
B. TECH (AI & DS) SEVENTH SEM
Tutorial 2
1. Exploratory Data Analysis (EDA)
o Importance of EDA
o Steps in EDA
o Tools for EDA
2. Statistical Analysis for Business Intelligence
o Descriptive Statistics
o Inferential Statistics
o Hypothesis Testing
3. Data Visualization Techniques and Tools
o Principles of Effective Data Visualization
o Common Visualization Techniques
o Popular Visualization Tools
4. Interactive Dashboards and Reports
o Designing Effective Dashboards
o Tools for Creating Interactive Dashboards
o Best Practices for Dashboard Design
1. Exploratory Data Analysis (EDA)
EDA is an approach to analyzing data sets to summarize their main characteristics, often using
visual methods. It is a crucial step in data analysis, helping to uncover underlying patterns, detect
anomalies, and test assumptions.
1.1 Importance of EDA
Understanding Data Structure: EDA helps in understanding the structure, distribution,
and patterns within the data.
Identifying Outliers and Anomalies: Detect unusual observations that could distort
analysis.
Feature Selection: Determine which variables are most important for analysis.
1.2 Steps in EDA
1. Data Collection and Cleaning: Load the data and clean it by handling missing values,
duplicates, and inconsistencies.
2. Data Profiling: Understand the data types, unique values, and summary statistics.
3. Univariate Analysis: Analyze each variable individually using visualizations like
histograms or box plots.
4. Bivariate and Multivariate Analysis: Explore relationships between multiple variables
using scatter plots, correlation matrices, and pair plots.
5. Outlier Detection: Identify and handle outliers using statistical techniques or visual
methods.
6. Feature Engineering: Create new features based on existing data to improve model
performance.
1.3 Tools for EDA
Python Libraries: Pandas, Matplotlib, Seaborn, Plotly
R Libraries: ggplot2, dplyr, tidyr
BI Tools: Tableau, Power BI, Qlik
2. Statistical Analysis for Business Intelligence
Statistical analysis involves collecting and analyzing data to identify patterns and trends. It
provides the foundation for evidence-based decision-making in business intelligence.
2.1 Descriptive Statistics
Measures of Central Tendency: Mean, median, and mode describe the central point of a
data set.
Measures of Dispersion: Range, variance, and standard deviation show how spread out
the data is.
Frequency Distributions: Understanding the distribution of categorical and numerical
data.
2.2 Inferential Statistics
Inferential statistics allow us to make predictions or inferences about a population based on a
sample of data.
Confidence Intervals: Estimating the range within which a population parameter lies.
Correlation and Regression Analysis: Assessing relationships between variables.
ANOVA (Analysis of Variance): Comparing means of multiple groups.
2.3 Hypothesis Testing
Hypothesis testing is a method for testing a claim or hypothesis about a parameter in a
population using sample data.
Null Hypothesis (H ): The default assumption (e.g., no difference between groups).
Alternative Hypothesis (H ): The assumption contrary to the null hypothesis.
P-Value: The probability of observing the data if the null hypothesis is true.
Types of Tests: t-test, chi-square test, z-test.
3. Data Visualization Techniques and Tools
Data visualization is the graphical representation of information and data. Effective
visualizations make complex data more accessible, understandable, and usable.
3.1 Principles of Effective Data Visualization
Clarity: Avoid clutter and ensure the visualization is easy to read.
Accuracy: Represent data correctly without distortion.
Simplicity: Use simple, intuitive designs to convey the message.
3.2 Common Visualization Techniques
Bar Charts: Compare quantities across categories.
Line Charts: Show trends over time.
Pie Charts: Display parts of a whole.
Scatter Plots: Show relationships between two variables.
Heatmaps: Visualize matrix data where values are represented by color.
Histograms: Show the distribution of a single variable.
3.3 Popular Visualization Tools
Tableau: User-friendly tool for creating interactive visualizations.
Power BI: Microsoft’s tool for creating reports and dashboards.
Google Data Studio: Free tool for building reports with Google data.
Plotly: Python-based library for creating interactive visualizations.
4. Interactive Dashboards and Reports
Interactive dashboards allow users to explore data dynamically and gain insights through visual
interactions. Reports present data in a structured format for decision-makers.
4.1 Designing Effective Dashboards
Identify the Audience: Tailor the dashboard to the needs of its users.
Focus on Key Metrics: Highlight important metrics that align with business goals.
Use Appropriate Visuals: Choose the right charts for the data being presented.
Ensure Interactivity: Enable filters, drill-downs, and hover effects for deeper insights.
4.2 Tools for Creating Interactive Dashboards
Tableau: Offers rich interactive features and supports various data sources.
Power BI: Integrates well with Microsoft products and supports real-time data.
Qlik Sense: Allows for associative data exploration and custom visuals.
Google Data Studio: Simple tool for integrating Google Analytics and other Google
services.
4.3 Best Practices for Dashboard Design
Keep it Simple: Avoid clutter and focus on essential information.
Use Consistent Design: Maintain uniform colors, fonts, and layout for a cohesive look.
Make it Accessible: Ensure the dashboard is easily accessible and navigable.
Test and Iterate: Gather feedback from users and refine the dashboard as needed.
